Lighting Question

[Jim Crenca]

My 94 KDX has a high output lighting coil that I primarily installed for the increased low end that has been mentioned here before. This weekend I made a wiring harness that would allow both elements of a stock bulb to light with the idea of a brighter headlight. On the bench it works great with either a DC supply (battery charger) or car battery, but on the bike, both elements are condsiderably dimmer than the stock set up of using just one. All connections are clean, low resistance, etc. Is this a voltage regulator problem or am I missing something. The coil (bought from Stealhy) is either a Moose or Electrex unit and is rated at approx 65 watts; the bulb is stock which I thought was about 25 watts (per element). Any ideas?

 

[Jim Crenca]

CC,
Where are you buddy, I know you or C.Dave have an answer for this one.

 

[canyncarvr]

BuyKawi says:

BULB,12V 30/30W

If you have the tail light in the circuit it's wanting 70 watts in just filaments..let alone wire/connection/switch losses, VR requirements and such. You will be just over the edge of what the Electrex coil can put out. It's sure not going to get close to max without a good bit of engine speed.

If things brighten up with the tail bulb out it's likely a watt shortage problem.

If you're running both filaments off the oem wiring, I'd not be surprised if the wire was too small for the current you're now expecting it to carry. The ground has to carry the current for both filaments.

You don't happen to be running the filaments in series? That won't work. Each 'hot' side of the filament (filaments don't HAVE a 'hot' side..until they're wired that way) needs to be directly connected to power. If you're using the oem socket, it would pretty darn tough to wire them in series...but just thought I'd ask. It's the word 'harness' that makes me wonder how you did it. A simple shunt soldered between the bulb input 'nubbins' would get done what you're trying to do...no harness required.

...don't know if the shell could handle that much heat, anyway.

All of this based on an 'H' model. I don't know but I'd doubt the '94 is that different.


So...what happens with the tail bulb out?

What does your 'harness' consist of?

What is the voltage in the circuit with the lights on? (The 'H' model is AC..not DC. Likely yours is AC, too). Should be 12VAC.

If the voltage drops when you light 'em up, you're either losing current someplace or don't have it to start with. If you measure different voltages at different places in the bike harness, you have a wiring/connection problem.

Where are I? Here very seldom. A fluke that I saw your post. I'll check back on it, though.

 

[Jim Crenca]

As always, thanks for the input.
The harness is each nubbin soldered to a short section of #12 wire; these are then "Y'd" at a bullet splice connector for a reasonably clean parallel conncection. It never occured to me theses bikes are AC; the filaments really light up with a 12 volt DC car battery though.
I've posted this question about heat before and I guess nobody ever wanted more output; I did check it with an infared heat gun (with a few minutes run time it got to 160 F.) and thought it might be OK; no soft plastic.
I also bypassed the wimpy stock ground wire at the shell and substituted a big old clip lead; no difference.
Also of note, the wires don't heat up when powered by bike, battery charger, or car battery so I'm surmising wire gauge is OK.
It must be current draw though as both filaments are REALLY dim and just one is plenty bright.
Tomorrow I'll try removing the tail light bulb and will make voltage drop measurements. I also want to check the schematic to see if the voltage regulator might be in series with output side of coil and possibly be the limiting factor.

Lastly, what forums do you follow.
I've got a tech question about helmet tests (from Motorcyclist magazine) that indicates that a DoT helmet may transmit less shock to a rider than a SNELL version. Have you heard about this?

 

[canyncarvr]

The VR is wired in parallel.

It does seem odd to me that with the Electrex coil rated at 65W, the 70W you're trying to suck out of it would turn the thing into 'REALLY dim.' Not as bright as it should be? Yeah. But not REALLY dim.

You'll get it.

re: dot/snell.

Nope.

re: Which? I don't think that's an allowed answer. ;) I do have email-to capability.

 

[Porter]

You'd be surprised at the amount of watts the stock taillight loop sucks up.I used a small automotive side marker light (1.5 watts, I think) and hard wired it into the rear light.Then I replaced the front bulb with a halogen 30W from flywheel weight guy and the difference is awesome! There is just about nill drop out at idle (brightness) and the beam is very cool.Yeah,I can actually ride at night now in the woods.

 

[matt-itude]

I like the halogens also. I dropped my buld one night getting ready to ride the next day (long night dont ask) the only bulb I could find was a halogen 30w out of a polaris explorer. Huge diffrence, much brighter, I think I will keep using them.

 

[Jim Crenca]

Both bulbs on at front makes tail light very dim.
Unhooking tailight doesn't make much difference for headlight.
There isn't much volatge drop at all; removing V.R from system makes no difference.
Measeured voltage was either 6 or 8 VAC; I forget which; there is a ton of ignition noise as one of my digital Fluke meters had ever changing readings.
A 30 watt halogen bulb using 1 element works best; tailight didn't effect this at all.
No reason to buy Electrex coil other than for slight performance increase as current output can not be a full 30 watts over stock.
Overall, the charging system is very cute and simple like it should be for a trail bike, but junk for anything serious.

 

[canyncarvr]

Need to take a couple steps back....

If removing the VR made no difference, there is a problem.

That's news for 'ya, huh?

The output of the coil with NO regulator/load can surpass 100VAC.

You don't have to actually remove the VR to see that...just look on the yellow wire with the light switch off.

If you don't see a LOT of volts on that wire, either your coil is shot, damaged, the joint at the yellow wire or the coil ground is bad.

Well...you could be a few gausses short on the rotor. That would be a bad thing...... You didn't happen to drop the rotor off your bench during the process, did'ja? :ohmy:

Natch...'testing' by hooking up the lights with the VR out of the circuit will (should) blow the things sky-high if you blip the throttle.

Your Fluke would indeed jump around a good bit...the output of the coil (sans VR) is pretty unstable.

Fix it! I wanna know what the problem is (was)! ;)

 

[Jim Crenca]

Thanks CC, I'll give it a try tonight.
I did have VR unhooked once 6 months ago and it did blow up the taillight bulb real good, lights were off though.
When I recently had VR unhooked ttaillight bulb was removed and headlights did not change; not enough current was my thought.
Rotor was handled gingerly, and I'm assuming that coil would either be good or bad as I remember it to be a 2 wire, series wound device.
Single headlight filament and tail light are nice and bright; it's 2 filaments that cause the problem.
I'll try a measurement as you described above at yellow wire and will also try "sacrificing" another taillight bulb with VR unhooked.
Do you think a bad ground / dirty connection would only cause 2 filament issue?

 

[canyncarvr]

Running dim lights with the VR out is either a problem...or you're sucking enough amps to clamp the output to a below-regulated level.

OK..it runs fine with one filament and the taillight. That's 40W (or so). You should have enough headroom with the Electrex coil to run another 30W fairly well at higher RPM. Maybe not as bright as the sum of the two filaments if individually lighted, but not real dim, either.

Idle output wouldn't be very good.

It would be handy to have a big rheostat. Hook the unregulated coil up to a fat load like that, crank the load up until you pull the source down to 12VAC. The measured current at that point will tell you exactly how many watts you gots.

Yes, the two types of loads aren't exactly the same (inductive, resistive and all that)...but would be good enough, I'd say.

Maybe a battery tester would work. Wouldn't take much of a load center to handle 70 or so watts. Heck, you could make one without too much trouble.

Oh yeah...the point....

That way you would KNOW what the output of the coil is.

re: Either good or bad?
No. Maybe they wound yours a couple layers short? Maybe it's wound in a sloppy fashion (not tight enough). Heck, rewind the thing yourself!

You could send the thing to RickyStator and have him check it out. Ask him to give you a wattage figure. He could fix it at the same time (given that you have a strong enough magnetic field).

What about this? We're assuming the coil is putting out 'as advertised'. If that assumption is incorrect, you're wasting your time trying to figure it out. There's nothing to figure!

Maybe you're just short on amps.

You still could be short on the magnet side of things. If you tested the coil on your bike, you would be reading the output of the coil based on its level of excitement (ooohhhh!!...) from your flywheel. It would be interesting to see if RickyStator got the same figure.


I'm looking forward to the facts-o-the-matter to be revealed! ;)

 

[Jim Crenca]

Oh this is getting so messy now; maybe 1994's are different than new bikes; here goes:

Slipped probes into back of connectors of wires going to VR

With VR unhooked, engine at fast idle, light switch OFF, measurement is 0 VAC
With VR hooked, fast idle, light switch ON, 11 to 13 VAC depending on RPM
With VR unhooked, fast idle, light switch ON, also 11 to 13 VAC

While I don't have a Kawasaki Special Meter as per manual, the VR tests bad; i.e. open from each wire to ground and about 5M ohm across leads

So if VR is history, how come it won't light both filaments
I started to check DC resistance across lighting coil but I don't have a spec; maybe it's time to call Electrex and get factory spec.
I suppose magnets could be wimpy, and I did not check brightness with both filaments at higher RPM (pretty stupid since I removed 1 filament wire from harness
Aren't you running an aftermarket light with approx 70 watts of lighting?

 

[canyncarvr]

Oh this is getting so messy now; maybe 1994's are different than new bikes; here goes:

Slipped probes into back of connectors of wires going to VR

With VR unhooked, engine at fast idle, light switch OFF, measurement is 0 VAC

That's right. The VR is out of the circuit with the light switch off.

With VR hooked, fast idle, light switch ON, 11 to 13 VAC depending on RPM

That sounds good, too.

With VR unhooked, fast idle, light switch ON, also 11 to 13 VAC

Oops. Not good. Well, I'm not sure about the 'fast idle' part, but the Electrex coil will put out a good 100VAC with the VR out of the circuit.

Wait a minute....

Recheck that with a known good ground. The VR is grounded to the frame. The two wires at the VR connector are hot unregulated IN and regulated OUT. With a meter connected between those two, then disconnecting the ground of the VR by unplugging it..you just hooked yourself up to a floater.

While I don't have a Kawasaki Special Meter as per manual, the VR tests bad; i.e. open from each wire to ground and about 5M ohm across leads

So if VR is history, how come it won't light both filaments
I started to check DC resistance across lighting coil but I don't have a spec; maybe it's time to call Electrex and get factory spec.

I don't know what the spec of the lighting coil is. Considering the wire size and the few number of wraps, it's going to be close to short, isn't it? I've not looked at mine..don't know.

re: If the VR is history...

Still need an unregulated VAC reading. With the VR unplugged, measure between the yellow wire to the switch or red wire after the switch (switch on) AND a good frame ground. Better unplug your lights, too..unless you want to buy new ones.

If you don't get lots'a volts, you have a wiring problem, a bad coil or a weak flywheel.

I suppose magnets could be wimpy, and I did not check brightness with both filaments at higher RPM (pretty stupid since I removed 1 filament wire from harness

??? The output of the coil is 'at speed' (whatever that means), not at idle. With 60W in the front and 10W in the back, it will be dim at idle.

I don't get the 'pretty stupid' part. I don't understand '..removed 1 filament wire from the harness..' part, either.

Aren't you running an aftermarket light with approx 70 watts of lighting?

I am currently running 60W (taillight has been out for some time..haven't bothered to replace it 'cuz I don't care) up front..one 35W, one 25W, one spot, one flood (I don't recall the exact º figures).

Yes, they dim at idle.

Check your VR ground, too. At least unthread/rethread the bolt. Maybe even run a separate (an additional) ground if there is a contact question.

 

[canyncarvr]

Well? Got it taken care of? :nod:

 

[Jim Crenca]

Lost the kickstarter nut this weekend but found the kicker.
Of course a standard 12 mm metric nut that fits a 19mm socket won't clear the knuckle so you have to grind it as the dealer doesn't have a replacement. Tonight I'll make some measurements; maybe it's the Loctitie on the VR threads acting as an insulator???

 

[razrbakcrzy]

Just wanted to let you know what I found out about the moose (electrex) high output coil I put in my 92 200. After running into the same problems, not enough power to run the baja lighting kit installed. I took it to the local alternator repair shop and they tested the output for me the results were as follows:

45 watts at idle
60 watts at normal riding rev's
65 at high rpms.

Now if you are running a 30w head light (2) 10w turn signals and a led break and tail light. It will dim severly and the turn signals will not work at idle. Not to mention the horn working.

The things just do not put out the power the are rated at.

Jim

 

[Mully]

CC,

Question. The power for the lights (from the VR) is DC??

Mully

 

[Benesesso]

Here's a real simple solution to all this. Go buy a $3.29 RECTIFIER from Radio Shack, and a small 12V battery. Hook them up, and your troubles are over if the rest of the system is OK. The rect. changes the AC to DC to charge the batt., which keeps the lights nice and bright at idle.

 

[m0rie]

Question. The power for the lights (from the VR) is DC??

Mully - The stock KDX VR is a Regulator not a Rectifier. The lighting coil puts out AC which can exceed 100v AC. The Regulator clamps the output to 12v AC so your lights don't go pop when the RPM's go up. Most dual sport kits come with a Rectifier that replaces the Regulator. The Rectifier changes the output of the coil from AC to DC and allows you to charge a DC battery, run DC lights etc...

-Maurice

 

[Mully]

Ok, that makes since.

Seems to me that he has created a voltage divider circuit some how or another. The front bulb filaments are in series instead of parallel????? Like maybe he has a frame ground that he is not aware of??

Mully

 

[Jim Crenca]

As it turns out, the problem for me was unreasonably high expectations and not mechanical in nature.
Voltage readings at the red wire with the VR unhooked is as follows:
28VAC at idle
50-60 VAC at normal riding RPM
98VAC - winging the throttle

My original bad reading at VR was a bad meter ground
So, with the VR hooked up and approx 60 watts of headlight load, it's really dim at idle, not as bright as a single element at running speed, but it does have a wider projection of light.
I just don't ride at high RPM in the dark so a single filament may be better.
IMO a rectifier and battery will not fix the 2 filament issue as it appears that there isn't enough current being generated at low RPM to satisfy the load.
I do remember feeling a bit more torque with the HO coil, so it's still an OK mod from that perspective.
CC, Any thoughts about making a Super HO coil?
Enough room to mount, enough HP to power, a stronger VR??

 

[canyncarvr]

You could make a second coil...link them together as is common on some other bikes.

Sure a good place to be careful. Can't have anything getting loose down there.

Also..any interruption/change to the flux field will effect the excitor coil. The change of the light coil does. (Interesting you found that to be true, btw. I've teased that idea in a number of different threads, haven't gotten a single bite on any of 'em. It does work, though. Cool, huh?) Would stink to end up with spark problems after you hung on 6lbs of coils.

You might try a different wattage. Can you get a 40W that fits that socket? I haven't looked.

That would work pretty good.

Maybe put some LEDs in your tail light. That would help, too.

Bet if you took the Electrex apart and rewound it nice and tight you could squeeze a tad more out of it.

HP to turn past the coil wouldn't be any big deal.

RickyStator is still an option, too. Give him a call. Ask him if you can reasonably expect better output from one of his coils.

BTW...there is a balance to the whole thing. Bigger wire gets you more current and more windings gets you more voltage..but it's not a matter of huge wire and gobs of windings is great. At some point 'thing's kind'a just don't work so well.

Glad to hear you at least know it's ok!

..right? You're good to go?

Good!!

 

[Jim Crenca]

I think I'll wait until fall for anymore lighting work as it's time to ride!
I did see a special bulb in the Dennis Kirk book; from PIA I think that is a super hi-tech, bright white, halogen with 60 watts of light output with a 30 watt current draw. Sound a little like a "face lift in a jar". Thanks for the input fellas. :cool: